Tellus: Series A, Dynamic Meteorology and Oceanography (Jan 2020)

On the contribution of internal climate variability to European future climate trends

  • T. Koenigk,
  • L. Bärring,
  • D. Matei,
  • G. Nikulin,
  • G. Strandberg,
  • E. Tyrlis,
  • S. Wang,
  • R. Wilcke

DOI
https://doi.org/10.1080/16000870.2020.1788901
Journal volume & issue
Vol. 72, no. 1
pp. 1 – 17

Abstract

Read online

Large historical and future ensemble simulations from the Max-Planck Institute and the Canadian Earth System Models and from CMIP5 have been analysed to investigate the uncertainty due to internal variability in multi-decadal temperature and precipitation trends over Europe. Internal variability dominates the uncertainties in temperature and precipitation trends in all seasons at 30-year time scales. Locally, seasonal 30-year temperature trends deviate up to ±3 °C from the ensemble mean trend. Thus, in the entire of Europe, local seasonal temperature changes until year 2050 from below −1 °C up to more than 4 °C are possible according to the model results. Up to 30% of all ensemble members show negative temperature trends until year 2050 in winter, up to 10% of the members in summer. Uncertainties of 30-year precipitation trends due to internal variability exceed the trends almost everywhere in Europe. Only in few European regions more than 75% of the members agree on the sign of the change until year 2050. In southern Sweden, minimum and maximum winter (summer) temperature trends in the next 30 years differ with up to 7 °C (5 °C) between individual members of the large model ensembles. Large positive temperature trends are linked to positive (negative) precipitation trends in winter (summer) in southern Sweden. This variability is attributed to the variability in large scale atmospheric circulation trends, mainly due to internal atmospheric variability. We find only weak linkages between the variability of temperature trends and the dominant decadal to multi-decadal climate modes. This indicates that there is limited potential to predict the multi-decadal variability in climate trends. The main findings from our study are robust across the large ensembles from the different models used in this study but at the local scale, the results depend also on the choice of the model.

Keywords